GeekRadar / The comparison desk

COMPARE.

GPU

NVIDIA

NVIDIA T400 4 GB

Graphics processor

Video memory: 4 GB^*
Memory bus: 64 bit^*
Rated memory bandwidth: 80 GB/s^*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture: Turing^*
Process node: 12 nm^*
GPU cores: 384^*
Graphics interface: PCIe 3.0 x16^*
Reported chip die area: 200 mm²^*
Transistors: 4,700 million^*
Graphics processor: TU117^*
Texture mapping units: 24^*
Raster output units: 16^*
Streaming multiprocessors: 6^*
Chip foundry: TSMC^*
Card or module format: Single-slot^*
Declared GPU TDP: 30 W^*
Suggested system PSU: 200 W^*
External power connectors: None^*
DirectX / feature level: 12 (12_1)^*
OpenGL: 4.6^*
OpenCL: 3.0^*
Vulkan: 1.4^*
CUDA compute capability: 7.5^*
Shader model: 6.9^*
GPU L2 cache: 1,024 KB^*

Memory & storage

Memory options: 4^*
Video memory: 4 GB^*
Video memory type: GDDR6^*
Memory bus: 64 bit^*
Rated memory bandwidth: 80 GB/s^*

Power specifications

Thermal design power: 30 W^*

System configuration

Detailed specifications

Specifications 1 · 28 fields

proc transistors: 4,700 million^*
proc die size: 200^* mm²
cores: 384^*
sm count: 6^*
base clock: 420^*
boost clock: 1,425^*
memory clock: 1,250^*
ram: 4^*
name: NVIDIA T400 4 GB^*
ram type: GDDR6^*
bus width: 64^*
bus interface: PCIe 3.0 x16^*
ram bandwidth: 80^* GB/s
fp16 performance: 2.189^*
fp32 performance: 1.0944^*
fp64 performance: 0.0342^*
tdp: 30^* W
suggested psu: 200^* W
slot width: Single-slot^*
api cuda: 7.5^*
is retail board: No^*
gpi value: 1.14^*
vendor: nvd^*
manufacturer: NVIDIA^*
arch name: Turing^*
card release date: 2021-05-06^*
proc foundry: TSMC^*
proc process size: 12^* nm

Specifications 2 · 38 fields

CUDA compute capability

7.5^*

DirectX / feature level

12 (12_1)^*

OpenCL

3.0^*

OpenGL

4.6^*

Shader model

6.9^*

Vulkan

1.4^*

Architecture

Turing^*

Graphics interface

PCIe 3.0 x16^*

Memory bus

64^* bit

Reported shader processors

384^*

Reported theoretical FP16 peak

2.189^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP32 peak

1.0944^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Reported theoretical FP64 peak

0.0342^* TFLOPS

Scope and conditions

Arithmetic density: Not stated^*

Declared L1 cache configuration

64 KB (per SM)^*

Declared GPU L2 cache

1024 KB^*

Reported display outputs

3x mini-DisplayPort 1.4a^*

Theoretical pixel fill rate

22.80 GPixel/s^*

External power connectors

None^*

Reported chip die area

200 mm²^*

Chip foundry

TSMC^*

Process node declaration

12 nm^*

Transistors

4,700 million^*

Graphics processor

TU117^*

Reported video memory

4^* GB

Rated memory bandwidth

80.00 GB/s^*

Video memory type

GDDR6^*

Raster output units

16^*

Reported shading units

384^*

Card or module format

Single-slot^*

Streaming multiprocessors

6^*

Suggested system PSU

200 W^*

2:4 tensor sparsity

No^*

Declared GPU TDP

30 W^*

Theoretical texture fill rate

34.20 GTexel/s^*

Texture mapping units

24^*

GPU cores

384^*

Video memory

4^* GB

Video memory type

GDDR6^*

* Provisional data